SicariusSicariiStuff/Phi-lthy4
Full fine-tune vs microsoft/phi-4. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 1/1, Norms 24/65, MLP 32/32, Attention 32/32, lm_head 1/1
4 derivatives checked
- bartowski/SicariusSicariiStuff_Phi-lthy4-GGUF: Same weights, converted
- mradermacher/Phi-lthy4-i1-GGUF: Quantization
- SicariusSicariiStuff/Phi-lthy4_GGUF: Quantization
- SicariusSicariiStuff/Phi-lthy4_iMatrix: Quantization
1 other repos hold byte-for-byte identical weight files.